Project Summary A growing number of reconstructed metabolic reaction networks have appeared in recent years. Such reconstruction can be converted into a mathematical format that in turn can be used to analyze the properties of the networks. Genome-scale analysis of network properties leads to understanding of their normal physiological functions and malfunction leading to pathophysiological states. The first part of this R01 program was focused on the analysis of possible steady state flux distributions of metabolic networks using extreme pathways. This analysis has proven useful for a number of purposes but is limited in scope. Fortunately, we have found random sampling of the solution space to be a viable and useful alternative to extreme pathways. During the second part of this program we propose to 1) develop uniform randomized sampling of the feasible steady state solution spaces, 2) to apply these algorithms to a series of biological and medical examples, and 3) develop time-scale decomposition of large- scale kinetic models. If the proposed program is successfully executed it will advance the state of the art of building genome-scale models to account for concentrations and to develop first-pass network-scale dynamic models. The availability of such genome-scale models would expand their scope of predictions and thus their applications to various biological and health care issues. PUBLIC HEALTH RELEVANCE: Project Narrative Public investment in DNA sequencing has led to the sequencing of entire genomes of a growing number of organisms. Scientists have determined how to assemble the information found in genomic sequences, along with data from the scientific literature, into the biochemical reaction networks that underlie cellular functions. This process is particularly advanced for metabolism; this proposal is focused on the mathematical description and analysis of metabolic functions in health and disease.